Hi everyone

First of all, thanks for this great forum with so many knowledgeable posters. I wonder if any of you can help me with the following questions.

Needing to accurately measure the black level on flat panel displays (LCD TVs and plasmas), I purchased an Extech 431025 light meter (which I believe is the same as the Robin RT24 and Tecpel DLM531). However, because the light meter only measures illuminance in lux or footcandles, I'm at a loss as to how to convert the readings to luminance (i.e. candela/square meter or nits or foot-Lambert).

Should I simply place the sensor directly on the TV screen? If so, how do I convert from lux/fc to cd/m2 or ft-L?

I understand that a spot-reading luminance meter like Konica-Minolta LS-100 will probably give the most accurate reading, but I just can't afford it at the moment. Are there any cheaper alternatives? Would a colorimeter like Display LT combined with HCFR software be enough for reliable use?

Many thanks in advance.

Best regards
John

Greetings

It took me 20 sec to find a web site that "converts lux to ft-L" and any other unit. :)

Regards

Hi

Thanks for your quick reply. From another thread, I found the formula ftL = lux * screen gain / 10.76.

Is that what you're referring to? Would the screen gain on flat panel displays be 1?

Thanks again.

Best regards
John

Greetings

Search convert lux to ft-L on any search engine ... google for instance ... it will lead to a site called efunda ... which gives you the conversion from any lux number you come up with into ft-L units.

(I purposely won't give you the link ;) )

Regards

caramonrun -- The cheapest meter, for measuring luminance, is the AEMC CA813 (about $160 from www.repaircalibration.com). IIRC, it is similar to the Extech meter you bought. You can find references to the CA813 in various calibration threads (not easy to find, though) which discuss the use of colorimeters (which are not designed to measure luminance, and are not accurate for that purpose).

@Michael TLV: I had a look at efunda, which suggested that 1 lux = 1 cd/m2 = 0.2918 ft-L according to efunda.com/units/convert_units.cfm?From=333. Surely that can't be right?

@CT_Wiebe: Measuring luminance is exactly the reason why I bought the meter, but I still can't figure out how to use it to measure black level on a flat panel display.

Let me give you an example. Using the Extech 431025 light meter, I recorded 3.2 lux when I placed the sensor directly on the screen of a properly calibrated 40-inch Samsung A650 LCD TV displaying a full-field black pattern. Could anyone please enlighten me as to how does that translate to cd/m2 or ft-L? Or am I doing something wrong here?

Many thanks again.

Best regards
John

Part of the problem is that a meter designed to read illuminance is probably going to have a diffuser attached to it. But you want to read luminance, which does not require a diffuser. You can try removing it it (if you can). Then the Lux reading will be much higher and you can multiply by 0.29 to get fL.

Anyway, short of having a luminance meter that you use to construct a mathematical conversion, that's what I would do.

@TomHuffman: I tried your suggestion, but the numbers don't tally up. By removing the diffuser (I assume it's a white plastic cover over the sensor), I recorded 16 lux (the sensor was placed directly on the LCD screen) which translates to 4.65 fL, an extremely bad black level which is not indicative of the excellent blacks on the Samsung A650.

Are there any cheaper alternatives? Would a colorimeter like Display LT combined with HCFR software be enough for reliable use?

John, from most everything I've read here and depending a lot of what you define as "reliable" use, a i1 D2/LT colorimeter would probably be fine for doing both color and light readings on your display. If you're looking for a more "accurate" reading of the light level at 0% stimulus (or any other stimulus) then you probably want a dedicated light meter like you purchased. You mentioned that the set has already been properly calibrated - was this using Avia/Avia II/other calibration disc or done by a professional? And are you just trying to get an accurate on/off contrast ratio measurement after calibration? If you can mention a bit more about what your ultimate goal is then folks may have additional advice on what you need, etc.

hope this helps,


--tom

@Tom: The set has been calibrated by myself using a i1 LT colorimeter and HCFR software, which incidentally gave a black level 0IRE reading of 0.07 cd/m2. Wishing to obtain the black level in an accurate fashion, I purchased a dedicated light meter.

Maybe I'm just dim, but I can't figure out how to use it to measure black-level luminance either in cd/m2 or ft-L. I've tried the suggested methods above, but the result deviates quite significantly from 0.07 cd/m2 (not that this is correct of course, but there shouldn't be such a big deviation if what I'm doing is right).

Here are my key questions:

1) The Extech 431025 measures illuminance in lux or foot-candle. How can I convert it to cd/m2?

2) Should I place the sensor directly on the LCD screen?

3) Should I remove the cosine corrector?

4) Or should I just accept that only a spot-reading luminance meter like a Konica-Minolta LS-100 is capable of measuring black level accurately?

Warmest regards
John

@Tom: The set has been calibrated by myself using a i1 LT colorimeter and HCFR software, which incidentally gave a black level 0IRE reading of 0.07 cd/m2. Wishing to obtain the black level in an accurate fashion, I purchased a dedicated light meter.

Maybe I'm just dim, but I can't figure out how to use it to measure black-level luminance either in cd/m2 or ft-L. I've tried the suggested methods above, but the result deviates quite significantly from 0.07 cd/m2 (not that this is correct of course, but there shouldn't be such a big deviation if what I'm doing is right).

Here are my key questions:

1) The Extech 431025 measures illuminance in lux or foot-candle. How can I convert it to cd/m2?

2) Should I place the sensor directly on the LCD screen?

3) Should I remove the cosine corrector?

4) Or should I just accept that only a spot-reading luminance meter like a Konica-Minolta LS-100 is capable of measuring black level accurately?OK, I didn't know that you had a colorimeter. Since removing the diffuser didn't work, keep it on. Then it is really simple.

1. Measure a 100% stimulus window with the i1 LT and record the value in cd/m2.
2. Measure the same window with your light meter using direct contact and record in lux.
Now you can calculate a cd/m2-lux conversion factor between the two meters.
3. Measure a black screen using the illuminance meter and convert the lux result using the conversion factor. Multiply that by 0.29 to get fL.

Since the i1 LT should be quite accurate with a 100% window, this should give you a very accurate result at low light levels where the lux meter is presumably more sensitive.

Hi Tom

I tried your method, but the result is again unsatisfactory:

1. 100% stimulus window: measured 120 lux using Extech 431025; 105 cd/m2 using i1 LT.

2. Conversion factor: 1.1429

3. 0% stimulus window: measured 2 lux using Extech 431025, which translates to 1.75 cd/m2, an even more erroneous result.

Incidentally, I found this post by D-nice: (http://www.avsforum.com/avs-vb/showpost.php?p=13630862&postcount=26)

Ok, I've been researching the 0.09lx rating on these LGs and found a converter that can convert lx to fL better than I can. 0.09lx actually equals 0.0084fL. Very impressive and is basically equal to the 11G Panasonics(0.009fL). Niether can beat the 8G Kuros with their 0.004fL rating....and dont even think about comparing these to the upcoming 9G Kros with 0.001fL and better minimum luminance levels.

... which suggests a conversion factor of 10.714. But is this true?

I also found your post (http://www.avsforum.com/avs-vb/showpost.php?p=13464288&postcount=707) stating the black level of the Panasonic 85U plasma to be 0.0071 fL. If you don't mind me asking, what instrument did you use to measure such low luminance?

Warmest regards
John

Hi Michael TLV, CT_Wiebe, TomHuffman and thomasl:

D-nice implied in another forum post (http://www.avsforum.com/avs-vb/showthread.php?p=14212506#post14212506) that the reason I've stopped receiving help in this thread is because of my overly aggressive questions and "thick" forum demeanor. If that's the case, I sincerely apologize... I didn't mean to offend anyone, I was just trying to find out exactly how to use an illuminance meter to measure black level.

Best regards
John

Greetings

So have you been able to verify that the measurement tool you have is fully functional or itself in need of recalibration?

Regards

Hi Tom

I tried your method, but the result is again unsatisfactory:

1. 100% stimulus window: measured 120 lux using Extech 431025; 105 cd/m2 using i1 LT.

2. Conversion factor: 1.1429

3. 0% stimulus window: measured 2 lux using Extech 431025, which translates to 1.75 cd/m2, an even more erroneous result.

Incidentally, I found this post by D-nice: (http://www.avsforum.com/avs-vb/showpost.php?p=13630862&postcount=26)



... which suggests a conversion factor of 10.714. But is this true?

I also found your post (http://www.avsforum.com/avs-vb/showpost.php?p=13464288&postcount=707) stating the black level of the Panasonic 85U plasma to be 0.0071 fL. If you don't mind me asking, what instrument did you use to measure such low luminance?I used the AEMC CA813.

BTW, I now question the results your meter is giving. The conversion factor I usually get is around 1 lux for every 0.45 cd/m2. It varies some (not much) between displays, which is why it is important to measure the offset. I just used on my CRT and I got 119 cd/m2 and 270 lux.

Hi Michael TLV & Tom:

Thanks for the suggestions. I'll see if I can get my hands on a replacement meter and report back.

Best regards
John

I used the AEMC CA813...Tom, any chance you could take cd/m^2 (i1 Pro) vs Lux (AEMC CA813) measurements of the entire gray scale in 11 steps, i.e. 0, 10%, 20%,... ..., 100% stimulus and present the raw results (or 21 steps if you have the time ;))?

It would be very interesting to see how the conversion factor between the two sensors change with changing luminance/illuminance.

These methods are all going to be a problem at the current black level.

All of these meters are using photopic correction curves when a scotopic or mesopic curve may be required. Scotopic vision occurs at less than 0.01 fL, photopic is at greater than 1 fL and mesopic is from 1 fL to 0.01 fL. The reason for a different response curve is that at very low light levels the eye switches from cones to rods. A correction factor for relative light adaptation would also be required. All of these factors may make it nearly impossible to develop an accurate way to measure perceptible black level without knowing the exact circumstances of display usage. For example if you are using a front projector in a dark room with a movie that is very dark you may come much closer to the scotopic curve while someone watching a plasma in a room with some light in it may never leave the photopic curve because they will not light adapt. I know when I am calibrating a front projector with a very dark test pattern for an extended time period the black level gets perceptually worse over time as my eyes adapt. If all you want to do is compare displays then go ahead and use a photopic curve for them all, but as black level drops and dark adaptation occurs the perceptual black level change may be nearly negated from increases in the eye's efficiency.

One approach would be to take a reading with a spectral device and calculate the luminance with a scotopic or mesopic curve and then compare that to the filter based meter and calculate a correction factor. With that factor you can then attempt a black level reading, but this is going to be difficult with many of the current low black level displays and the sensitivity of most light meters. A device like this should have the necessary sensitivity http://unihedron.com/projects/darksky/, but it will require compensation to correct for its calculation method.

Tom, any chance you could take cd/m^2 (i1 Pro) vs Lux (AEMC CA813) measurements of the entire gray scale in 11 steps, i.e. 0, 10%, 20%,... ..., 100% stimulus and present the raw results (or 21 steps if you have the time ;))?

It would be very interesting to see how the conversion factor between the two sensors change with changing luminance/illuminance.

Sure. That's easy to do.

http://home.comcast.net/%7Etlhuffman/gamma/data1.gif

Here is the raw data. Notice how the D2/CA813 conversion factor stays ruler flat all the way down to black. On the other hand, the i1Pro begins to change at .4 stim and by .1 stim has completely fallen apart. I didn't get a reading at black at all.

The result is that the i1Pro gamma reading is pretty much useless below at .3 stim and below. Interestingly, the D2 had small problems at the high end as well.

http://home.comcast.net/%7Etlhuffman/gamma/chart.gif

Here are the cd/m2 and fL readings from the CA813 using the conversion factors. Notice that the contrast reading is identical to what I got from the D2. Because the i1Pro couldn't read black, no contrast was available.

http://home.comcast.net/%7Etlhuffman/gamma/data2.gif

These readings are all from the Panasonic TH-PZ5085U plasma.

Cheers for that Tom, interesting results. What software did you use to take the measurements?

Thanks, Jeff, for your elaboration (if that is the correct English term?). Since the eye functions between ~3x10^-7 cd/m^2 to ~3x10^4 cd/m^2 it'll be a long time until there's an instrument equally capable. I didn't find any numbers on when the "Sky Quality Meter" becomes saturated as luminance increase... will it only function for reading the black level?